ReadColumnMatrix::ReadColumnMatrix(string df) : distFile(df){
- successOpen = openInputFile(distFile, fileHandle);
+ successOpen = m->openInputFile(distFile, fileHandle);
+ sim = false;
}
+/***********************************************************************/
+
+ReadColumnMatrix::ReadColumnMatrix(string df, bool s) : distFile(df){
+
+ successOpen = m->openInputFile(distFile, fileHandle);
+ sim = s;
+}
/***********************************************************************/
-void ReadColumnMatrix::read(NameAssignment* nameMap){
+int ReadColumnMatrix::read(NameAssignment* nameMap){
try {
+
+ string firstName, secondName;
+ float distance;
+ int nseqs = nameMap->size();
+ DMatrix->resize(nseqs);
+ list = new ListVector(nameMap->getListVector());
- string firstName, secondName;
- float distance;
- int nseqs = nameMap->size();
+ Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs);
- list = new ListVector(nameMap->getListVector());
-
- Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs);
+ int lt = 1;
+ int refRow = 0; //we'll keep track of one cell - Cell(refRow,refCol) - and see if it's transpose
+ int refCol = 0; //shows up later - Cell(refCol,refRow). If it does, then its a square matrix
+
+ //need to see if this is a square or a triangular matrix...
- int lt = 1;
- int refRow = 0; //we'll keep track of one cell - Cell(refRow,refCol) - and see if it's transpose
- int refCol = 0; //shows up later - Cell(refCol,refRow). If it does, then its a square matrix
+ while(fileHandle && lt == 1){ //let's assume it's a triangular matrix...
+
+
+ fileHandle >> firstName; m->gobble(fileHandle);
+ fileHandle >> secondName; m->gobble(fileHandle);
+ fileHandle >> distance; // get the row and column names and distance
+
+ if (m->debug) { cout << firstName << '\t' << secondName << '\t' << distance << endl; }
+
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
- //need to see if this is a square or a triangular matrix...
- while(fileHandle && lt == 1){ //let's assume it's a triangular matrix...
+ map<string,int>::iterator itA = nameMap->find(firstName);
+ map<string,int>::iterator itB = nameMap->find(secondName);
+
+ if(itA == nameMap->end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
+ if(itB == nameMap->end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
+
+ if (distance == -1) { distance = 1000000; }
+ else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert.
- fileHandle >> firstName >> secondName >> distance; // get the row and column names and distance
-
- if(nameMap->count(firstName)==0){
- cerr << "AError: Sequence '" << firstName << "' was not found in the names file, please correct\n";
+ if(distance < cutoff && itA != itB){
+ if(itA->second > itB->second){
+ PDistCell value(itA->second, distance);
+
+
+ if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
+ refRow = itA->second;
+ refCol = itB->second;
+ DMatrix->addCell(itB->second, value);
+ }
+ else if(refRow == itA->second && refCol == itB->second){
+ lt = 0;
+ }
+ else{
+ DMatrix->addCell(itB->second, value);
+ }
}
- if(nameMap->count(secondName)==0){
- cerr << "AError: Sequence '" << secondName << "' was not found in the names file, please correct\n";
+ else if(itA->second < itB->second){
+ PDistCell value(itB->second, distance);
+
+ if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
+ refRow = itA->second;
+ refCol = itB->second;
+ DMatrix->addCell(itA->second, value);
+ }
+ else if(refRow == itB->second && refCol == itA->second){
+ lt = 0;
+ }
+ else{
+ DMatrix->addCell(itA->second, value);
+ }
}
+ reading->update(itA->second * nseqs);
+ }
+ m->gobble(fileHandle);
+ }
+
+ if(lt == 0){ // oops, it was square
+
+ fileHandle.close(); //let's start over
+ DMatrix->clear(); //let's start over
+
+ m->openInputFile(distFile, fileHandle); //let's start over
+
+ while(fileHandle){
+ fileHandle >> firstName; m->gobble(fileHandle);
+ fileHandle >> secondName; m->gobble(fileHandle);
+ fileHandle >> distance; // get the row and column names and distance
- if (distance == -1) { distance = 1000000; }
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
+
+ map<string,int>::iterator itA = nameMap->find(firstName);
+ map<string,int>::iterator itB = nameMap->find(secondName);
- if(distance < cutoff && nameMap->get(firstName) != nameMap->get(secondName)){
- if(nameMap->get(firstName) > nameMap->get(secondName)){
- PCell value(nameMap->get(firstName), nameMap->get(secondName), distance);
+ if(itA == nameMap->end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
+ if(itB == nameMap->end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
- if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
- refRow = nameMap->get(firstName);
- refCol = nameMap->get(secondName);
- D->addCell(value);
- }
- else if(refRow == nameMap->get(firstName) && refCol == nameMap->get(secondName)){
- lt = 0;
- }
- else{
- D->addCell(value);
- }
- }
- else if(nameMap->get(firstName) < nameMap->get(secondName)){
- PCell value(nameMap->get(secondName), nameMap->get(firstName), distance);
+ if (distance == -1) { distance = 1000000; }
+ else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert.
- if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
- refRow = nameMap->get(firstName);
- refCol = nameMap->get(secondName);
- D->addCell(value);
- }
- else if(refRow == nameMap->get(secondName) && refCol == nameMap->get(firstName)){
- lt = 0;
- }
- else{
- D->addCell(value);
- }
- }
- reading->update(nameMap->get(firstName) * nseqs);
+ if(distance < cutoff && itA->second > itB->second){
+ PDistCell value(itA->second, distance);
+ DMatrix->addCell(itB->second, value);
+ reading->update(itA->second * nseqs);
}
- gobble(fileHandle);
+
+ m->gobble(fileHandle);
}
+ }
+
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
+
+ reading->finish();
+ fileHandle.close();
- if(lt == 0){ // oops, it was square
- fileHandle.close(); //let's start over
- D->clear(); //let's start over
-
- openInputFile(distFile, fileHandle); //let's start over
+ list->setLabel("0");
+
+ return 1;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ReadColumnMatrix", "read");
+ exit(1);
+ }
+}
+/***********************************************************************/
- while(fileHandle){
- fileHandle >> firstName >> secondName >> distance;
+int ReadColumnMatrix::read(CountTable* countTable){
+ try {
+
+ string firstName, secondName;
+ float distance;
+ int nseqs = countTable->size();
+
+ DMatrix->resize(nseqs);
+ list = new ListVector(countTable->getListVector());
+
+ Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs);
+
+ int lt = 1;
+ int refRow = 0; //we'll keep track of one cell - Cell(refRow,refCol) - and see if it's transpose
+ int refCol = 0; //shows up later - Cell(refCol,refRow). If it does, then its a square matrix
+
+ //need to see if this is a square or a triangular matrix...
+
+ while(fileHandle && lt == 1){ //let's assume it's a triangular matrix...
+
+
+ fileHandle >> firstName; m->gobble(fileHandle);
+ fileHandle >> secondName; m->gobble(fileHandle);
+ fileHandle >> distance; // get the row and column names and distance
+
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
+
+ int itA = countTable->get(firstName);
+ int itB = countTable->get(secondName);
+
+ if (m->control_pressed) { exit(1); }
+
+ if (distance == -1) { distance = 1000000; }
+ else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert.
- if(nameMap->count(firstName)==0){
- cerr << "BError: Sequence '" << firstName << "' was not found in the names file, please correct\n";
+ if(distance < cutoff && itA != itB){
+ if(itA > itB){
+ PDistCell value(itA, distance);
+
+
+ if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
+ refRow = itA;
+ refCol = itB;
+ DMatrix->addCell(itB, value);
}
- if(nameMap->count(secondName)==0){
- cerr << "BError: Sequence '" << secondName << "' was not found in the names file, please correct\n";
+ else if(refRow == itA && refCol == itB){
+ lt = 0;
}
-
- if (distance == -1) { distance = 1000000; }
-
- if(distance < cutoff && nameMap->get(firstName) > nameMap->get(secondName)){
- PCell value(nameMap->get(firstName), nameMap->get(secondName), distance);
- D->addCell(value);
- reading->update(nameMap->get(firstName) * nseqs);
+ else{
+ DMatrix->addCell(itB, value);
+ }
+ }
+ else if(itA < itB){
+ PDistCell value(itB, distance);
+
+ if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol...
+ refRow = itA;
+ refCol = itB;
+ DMatrix->addCell(itA, value);
+ }
+ else if(refRow == itB && refCol == itA){
+ lt = 0;
+ }
+ else{
+ DMatrix->addCell(itA, value);
}
-
- gobble(fileHandle);
}
+ reading->update(itA * nseqs);
}
- // else if(lt == 0){
- // while(fileHandle){
- // fileHandle >> firstName >> secondName >> distance;
- //
- // if(nameMap->count(firstName)==0){
- // cerr << "CError: Sequence '" << firstName << "' was not found in the names file, please correct\n";
- // }
- // if(nameMap->count(secondName)==0){
- // cerr << "CError: Sequence '" << secondName << "' was not found in the names file, please correct\n";
- // }
- // if (distance == -1) { distance = 1000000; }
+ m->gobble(fileHandle);
+ }
+
+ if(lt == 0){ // oops, it was square
+
+ fileHandle.close(); //let's start over
+ DMatrix->clear(); //let's start over
+
+ m->openInputFile(distFile, fileHandle); //let's start over
+
+ while(fileHandle){
+ fileHandle >> firstName; m->gobble(fileHandle);
+ fileHandle >> secondName; m->gobble(fileHandle);
+ fileHandle >> distance; // get the row and column names and distance
+
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
+
+ int itA = countTable->get(firstName);
+ int itB = countTable->get(secondName);
+
+
+ if (m->control_pressed) { exit(1); }
+
+ if (distance == -1) { distance = 1000000; }
+ else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert.
+
+ if(distance < cutoff && itA > itB){
+ PDistCell value(itA, distance);
+ DMatrix->addCell(itB, value);
+ reading->update(itA * nseqs);
+ }
+
+ m->gobble(fileHandle);
+ }
+ }
- // if(distance < cutoff && (*nameMap)[firstName].second < (*nameMap)[secondName].second){
- //// cout << (*nameMap)[secondName] << ' ' << (*nameMap)[firstName] << ' ' << distance << endl;
- // D->addCell(Cell((*nameMap)[secondName].second, (*nameMap)[firstName].second, distance));
- // reading->update((*nameMap)[secondName].second * nseqs);
- // }
- //
- // gobble(fileHandle);
- // }
- // }
- reading->finish();
- fileHandle.close();
-
- list->setLabel("0");
-
+ if (m->control_pressed) { fileHandle.close(); delete reading; return 0; }
+
+ reading->finish();
+ fileHandle.close();
+
+ list->setLabel("0");
+
+ return 1;
+
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the ReadColumnMatrix class Function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "ReadColumnMatrix", "read");
exit(1);
}
- catch(...) {
- cout << "An unknown error has occurred in the ReadColumnMatrix class function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
-
}
/***********************************************************************/
-
-ReadColumnMatrix::~ReadColumnMatrix(){
- //delete D;
- //delete list;
-}
-
+ReadColumnMatrix::~ReadColumnMatrix(){}
+/***********************************************************************/